Molecular markers in plant systematics and population 

biology

7. DNA sequencing (cpDNA)

Tomáš Fér

tomas.fer@natur.cuni.cz

DNA sequencing

• detection the order of nucleotides in a DNA strand

…ATATATAGGCAAGGAATCTCTATTATTAAATCATT…

• use the information to model evolutionary processes

• make hypothesis about similarity and relationships among taxa

Sequencing principle

• PCR with a primer pair• amplification of the target region

• cycle sequencing (dideoxy, Sanger)• use of one primer only• dNTP as well as ddNTP are present in the mixture• produce fragments differing exactly by one base

• electrophoretic separation of fragments in the gel• automated sequencer

2´, 3´‐ dideoxy NTPs

dTTP ddTTP

3´-CTGGACTGCA-5´5´-GACCT

Cycle sequencing

5´-ATGCATGC-3´3´-TACG-5´ primer

template

ddGTP

ATGCATGCGTACG

ddCTP

ATGCATGCCGTACG

ddATP

ATGCATGCACGTACG

ATGCATGC

ddTTP

TACGTACG

Automated sequencer

ABI (Applied Biosystems) – gel, capillary systems (up to 96) 

Genome structure• genetic information – order of nucleotides (ACGT)

• coding regions – exons – conserved• non‐coding regions – introns, spacers – variable

• nuclear, chloroplast and mitochondrial genome

5´ UTR       exon 1      intron       exon 2    3´ UTR       intergenic spacer exon

GENE 1 GENE 2

Sequence evolutionary rate

Types of variability in DNA sequences

5bp indel point mutations (SNPs)

Chloroplast genome

• many genes are single‐copy (only 1 copy in the whole genome)

• conserved evolution of the chloroplast genome• disadvantage when studying intraspecific or population variability

• many conserved regions can be used as priming sites

• structural rearrangements of chloroplast genome• mainly on larger evolotionary scale

• inversion – e.g., 30kb inversion differentiates bryophytes and higher plants

• extensive deletions

• loss of specific genes and intrones

• chloroplast capture• chloroplast transfer from one species to another by introgression

• can influence phylogeny in a wrong way (when not recognized)

Chloroplast genome

• 4 rRNAs

• 30‐11 tRNAs

• 21 ribosomal proteins (rps)

• 4 RNA polymerase subunits (rpo)

• 28 thylakoid proteins (ps)

• rbcL (large RuBisCO subunit)

• 11 proteins similar to NADH (ndh)

Frequently sequenced cpDNA regions

+ many others…

rbcL

• gene for large subunit of ribuloso‐1,5‐bisphosphate‐carboxylase/oxygenase(RUBISCO)

• 1,428, 1,431 or 1,434 bp in length – indelsare extremely rare

• one of the first sequenced genes• very conserved, systematics at family or generic level, in some groups at species level

atpB• gene coding beta subunit of ATP synthase

• 1,497 bp in length, indels not found

• similar use as rbcL

• codes a subunit of chloroplast NADH‐dehydrogenase

• 2,233 bp in length (tobacco)

• about 2x more substitutions then rbcL

• for generic level

ndhF

matK

• gene coding maturase (splicing of plastid genes)

• about 1,550 bp in length – low number of indels

• systematics at family and generic level

trnL intron and spacer between trnL and trnF

• tRNA genes – secondary structure• accumulation of insertions/deletionswith the same rate as nucleotide substitutions

• alignment problems, especially in distant organisms (sometimes already at family level)

• suitable for systematics of (closely) related species

trnLUAA5´exon

trnLUAA3´exon trnFGAA

intron spacer

atpB‐rbcL

• spacer of about 900‐1,000 bp in length

• systematics at family and generic level

Variable non‐coding cpDNA regionsShaw et al. (2005): The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA 

sequences for phylogenetic analysis. Am. J. Bot. 92: 142‐166

trnH‐psbApsbA‐trnK

rpS16

trnS‐trnG

rpoB‐trnC

trnD‐trnT

trnC‐ycf6ycf6‐psbMpsbM‐ trnD

trnS‐trnfM

trnS‐rpS4rpS4‐trnTtrnT‐trnLtrnL‐trnF

5´rpS12‐rpL20

psbB‐psbH

rpL16

trnG‐trnG

Variable non‐coding cpDNA regions

Shaw et al. (2007): Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. Am. J. Bot. 94: 275–288.• another 13 regions

Shaw et al. (2014): Chloroplast DNA sequence utility for the lowest phylogenetic and phylogeographic inferences in angiosperms: The tortoise and the hare IV. Am. J. Bot. 101: 1987–2004.• top 13 regions within each major evolutionary lineage

Use of chloroplast sequences

• phylogeny of angiosperms

• among‐species relationships within a genus

• within‐species phylogeography (haplotype definition)

• hybridization – inference of the maternal taxon (individual) – cpDNA maternaly inherited in angiosperms

Jansen et al. (2007)81 plastid genes76,583 nucleotides

Angiosperm phylogeny

Relationships among species

CapsicumatpB‐rbcL spacerWalsh & Hoot (2001)

Phylogeography

Arabis alpinatrnL‐trnFKoch et al. 2006

Inter‐specific hybridization

PersicariamatK, psbA‐trnH, trnL‐trnFversus ITSKim & Donoghue 2008

incongruence between cpDNA and nDNA

Data analysis• multiple alignment

• construction of phylogenetic tree• distance methods

• maximum parsimony (MP)

• maximum likelihood (ML)

• Bayesian inference (BI)

S206 ATATATATATAGGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

S207 ATATATATA--GGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

S208 ATATATATA--GGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

S209 ATATATATA--GGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

S210 ATATATATA--GGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

S0G3 ATATATATA--GGCAAGGAATCTCTATTATTAAATCATTTAGAATCCATA

TL ATATATATATAGGCAAGGAATCTCTATTATTAAATCATTCATAATTCATA

Maximum parsimony (MP)

• cladistic method• search for the simplest tree (most parsimonioustree)

• i.e., tree in which the evolution is explained by minimum number of substitutions

• software• PAUP * Phylogenetic Analysis Using Parsimony(* and other methods)

• TNTTree Analysis Using New Technology

Maximum likelihood (ML)

• search for tree with the highest probability (likelihood – L)

• probability that observed sequences evolved under given tree topology (and under given evolutionary model)

• software GARLI, PhyML, RAxML, PAML…

Evolutionary models for DNA sequences

• models for sequence changes

• parameters• base frequences• substitution types (transitions, transversions)• heterogeneity in substitution rates (G)• proportion of invariant sites (I)

A

T C

G purines

pyrimidines

transition

transversion

Substitution modelsJC (Jukes‐Cantor 1969)– same substitution rates– same base frequencies

A

CT

G

a

aa

a

a

a

A

CT

G

a

aa

a

a

a

A

CT

G

a

aa

b

b

a

A

CT

G

d

ce

f

a

b

K2P (Kimura 2 parameter 1980)– two different substitution rates– same base frequencies

F81 (Felsenstein 1981)– same substitution rates– different base frequencies HKY (Hasegawa, Kishino & Yano 1985)

– two different substitution rates– different base frequencies

GTR (General time‐reversible model)(Tavaré et al. 1986)– six different substitution rates– different base frequencies

Increasing

 num

ber of param

eters

b

A

CT

G

a

aa

b

a

Which model to select?

• MODELTEST: A tool to select the best‐fit model of nucleotide substitution (Posada et al.)

• testing different models – selecting the simpliestthat sufficiently explain the data using• hierarchical likelihood ratio tests (hLRTs)

• Akaike information criterion (AIC)

• jModelTest2 (https://code.google.com/p/jmodeltest2/)

Saturation• signal and noise in the data

• corrected versus uncorrecteddistance

• skewness (g1‐statistics), ISS

Gontcharov & Melkonian 2008

Molecular clock

• strict (global)• clocklike evolution

• local

• relaxed clocks• autocorrelated (closely related taxa have similar mutation rates)

• uncorrelated (lognormal, exponential)

• calibration• substitution rates from another study or generally assumed rate (e.g., for cpDNA)

• fossils

• biogeography

• software• BEAST (Bayesian), r8s (non‐parametric rate smoothing, penalized likelihood), … 

http://sydney.edu.au/science/biology/meep/documents/Workshop_Lecture4.pdf

Estimates of divergence times(BEAST – Bayesian Evolutionary Analysis Sampling Trees)

Antonelli A. (2009): Have giant lobelias evolved several times independently? Life form shifts and historical biogeography of the cosmopolitan and highly diverse subfamily Lobelioideae (Campanulaceae). BMC Biol. 7:82

Gene banks – databases of sequences

• GenBankNational Centre for Biotechnology Information (NCBI)http://www.ncbi.nlm.nih.gov/

• EMBLEuropean Bioinformatics Institute (EBI) http://www.ebi.ac.uk/embl/

GenBank example

Population studySanz M. et al. (2014): Southern isolation and northern long‐distance dispersal shaped the phylogeography of the widespread, but highly disjunct, European high mountain plant Artemisia eriantha (Asteraceae). Botanical Journal of the Linnean Society 174: 214–226.

Systematic study

Renner S.S. (2004): A chloroplast phylogeny of Arisaema (Araceae) illustrates Tertiary floristic links between Asia, North America, and East Africa. American Journal of Botany 91(6): 881–888

Literature

Soltis D.E. & al. [eds.] (1998): Molecular systematics of plants.II. DNA sequencing.

Hollingsworth & al. [eds.] (1999): Molecular systematics and plant evolution.

Hall B.G. (2001): Phylogenetic trees made easy.

Felsenstein J. (2004): Inferring phylogenies.

Lemey P. & al. [eds.] (2009): The phylogenetic handbook. 2nd ed.

Wiley E.O. & Lieberman B.S. (2011): Phylogenetics. Theory and Practice of Phylogenetic Systematics. 2nd ed.

Wheeler W. C. (2012): Systematics. A course of lectures.

Drummond et al. (2009): Relaxed phylogenetics and dating with confidence. PLoS Biol 4(5): e88